As is well known, human vision is one of senses for obtaining information of surrounding environments, and can recognize a position of an object, and farness and closeness through two eyes. That is, visionary information inputted through two eyes is synthesized to one distance information to allow a human to freely activate. A stereo camera system may be the one used to realize a visionary structure into a machine.
The stereo camera system performs a stereo matching relative to an image obtained by using two cameras, where the stereo camera system obtains a depth image using binocular disparity of two cameras in the stereo matching process. Depth image data is used for recognition and a 3D image can be generated using the depth image data.
Meantime, it is difficult for a stereo camera to recognize an image under a low light level, thereby creating a problem in obtaining a depth image.